[2022/10/24 19:41:47] ppocr ERROR: lang latin is not support, we only support dict_keys(['en', 'ch']) for layout models

[JonhSilver]

Why you declaring PPStructure support for 80+ languages?

[2022/10/24 19:41:47] ppocr ERROR: lang latin is not support, we only support dict_keys(['en', 'ch']) for layout models

[OpenOneV]

They should declare not-fully supported.

OCR only models might support more languages. But my Latin OCR result gives En letters.

paddleocr --lang latin --image_dir C:\Users\a_user\Pictures\OIP-C.jpeg

paddleocr --lang sdfsdfse --image_dir C:\Users\a_user\Pictures\OIP-C.jpeg

AssertionError: param lang must in dict_keys(['ch', 'en', 'korean', 'japan', 'chinese_cht', 'ta', 'te', 'ka', 'latin', 'arabic', 'cyrillic', 'devanagari']), but got sdfsdfse

[JonhSilver]

@OpenOneV paddleocr Latin language is wrong, as it used if i want Lithuanian and also return only latinian characters, if i change latin_dic tu lithuanian symbols it stop works at all

PPStructure is only accept en and ch

[OpenOneV]

@JonhSilver https://github.com/PaddlePaddle/PaddleOCR/issues/4415 Is this page helpful?

Assume you have some translator plugin on your browser.

[JonhSilver]

@JonhSilver #4415 Is this page helpful?

Assume you have some translator plugin on your browser.

ofcource no, also i dont know chinian language

[OpenOneV]

This is translated page img of #4415.

I really suggest you using browser like Edge/Chrome for built-in webpage translation, if you wish to proceed into this project. This PaddleOCR project is primarily maintained by Chinese programmers. Many important issues are written in Chinese. So with such a translator, you can go deeper. Meanwhile, a trans-language search engine is in great need for "international projects" primarily maintained by non-English speaking programmers.